Deep Learning-Based Constellation Optimization for Physical Network Coding in Two-Way Relay Networks

TL;DR

This paper introduces a deep learning approach to optimize constellations in two-way relay networks with physical-layer network coding, achieving higher sum rates and facilitating soft-decision decoding.

Contribution

It presents a novel DNN-based modulation and demodulation scheme that directly minimizes cross entropy, improving performance over traditional methods.

Findings

Significant sum rate improvements demonstrated through simulations

DNN-based demodulator outputs enable easy integration with soft-decision decoding

Method extends to higher-level constellations with minimal modifications

Abstract

This paper studies a new application of deep learning (DL) for optimizing constellations in two-way relaying with physical-layer network coding (PNC), where deep neural network (DNN)-based modulation and demodulation are employed at each terminal and relay node. We train DNNs such that the cross entropy loss is directly minimized, and thus it maximizes the likelihood, rather than considering the Euclidean distance of the constellations. The proposed scheme can be extended to higher level constellations with slight modification of the DNN structure. Simulation results demonstrate a significant performance gain in terms of the achievable sum rate over conventional relaying schemes. Furthermore, since our DNN demodulator directly outputs bit-wise probabilities, it is straightforward to concatenate with soft-decision channel decoding.